Humanitarian programming in fragile economies often use unconditional cash transfers (UCTs) to offset food insecurity. However, there is an increasing focus on using cash transfers to boost household incomes beyond the short-term through micro-enterprise start-up and growth. This paper conducts a randomised control trial to measure the impact of three different sizes of business grants against UCT in Somalia. We find that giving the same amount of money as a lump sum business grant results in higher likelihood of business ownership and income compared with UCT in the short run (3-4 months after the transfers). However, the impacts are larger and persist 3 years later only for those who received larger amount of grants. The results indicate our 'medium'-sized grant being more cost-effective. © 2022 The Author(s).